High density lipoprotein levels have been inversely correlated with the incidence and prevalence of coronary heart attacks. High denisty lipoprotein levels are higher in women, blacks, physically active and among individuals who consume alcoholic beverages. Alcohol consumption decreases testosterone levels in blood and high doses are assoicated with testicular disease. Individuals with low testosterone levels have been noted to have elevated high density lipoprotein. We hypothesize that the elevated high density lipoprotein associated with alcohol consumption is secondary to decrease plasma testosterone levels. Plasma testosterone will be measured among men participating in the Multiple Risk Factor Intervention Trial (MRFIT) project. A history of alcohol consumption, physical activity, weight change and dietary intake will also be determined. The levels of high density lipoprotein, low density lipoprotein, very low density lipoprotein cholesterol will be related to the plasma testosterone levels, alcohol consumption, physical activity, obesity, weight change and nutritional intake.